Every selfdriven.consulting engagement runs on KERI and ACDC — the cryptographic identity infrastructure that makes agent authority, audit trails, and human accountability provable by architecture, not just policy.
The core problem with AI-agent consulting is trust: how do you know who authorised the agent? How do you know an agent's findings haven't been tampered with? How do you know the human claiming accountability actually has it?
KERI (Key Event Receipt Infrastructure) and ACDC (Authentic Chained Data Containers) solve this at the infrastructure level — not through policy or contractual promises, but through cryptographic proof that can be independently verified.
Traditional identity systems — OAuth, SAML, Active Directory — require a central authority to vouch for identities. If that authority fails, is breached, or becomes adversarial, the system collapses. KERI identifiers are self-certifying: the proof of identity is in the key events themselves, not in any registry or authority.
This means every conductor and agent identity in a selfdriven engagement is verifiable by anyone, at any time, without asking permission of any central party.
Eight concepts underpin the KERI/ACDC stack. You don't need to understand all of them to benefit — but knowing what they are makes the trust model concrete.
The stack operates in three layers — each building on the previous. Together they make claims at every level of an engagement cryptographically provable.
For technical teams who want to understand the stack in detail before integration.